1_LogReg_XGB_MLP_CLF_Classifier.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Sahel-Eskandar/Project-Notebooks/blob/main/1_LogReg_XGB_MLP_CLF_Classifier.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "RfFtonm8VyQl"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.model_selection import KFold\n",
        "from sklearn.model_selection import StratifiedKFold\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.metrics import roc_auc_score\n",
        "from sklearn.metrics import roc_curve\n",
        "from sklearn.metrics import accuracy_score, confusion_matrix, precision_score, recall_score, classification_report\n",
        "from sklearn.dummy import DummyClassifier\n",
        "from sklearn import svm\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from xgboost import XGBClassifier\n",
        "from sklearn.neural_network import MLPClassifier\n",
        "from sklearn.metrics import roc_curve\n",
        "from sklearn.metrics import auc\n",
        "from sklearn.ensemble import IsolationForest\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "%matplotlib inline "
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Data Prepration"
      ],
      "metadata": {
        "id": "iVacmLdYkESs"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9CUsv11XVyQm"
      },
      "outputs": [],
      "source": [
        "path_y = Output + 'df_y.csv'\n",
        "path_X = Output + 'df_X.csv'"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "a1wVoG-HVyQn"
      },
      "outputs": [],
      "source": [
        "X = pd.read_csv(path_X)\n",
        "y = pd.read_csv(path_y)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bMBYXfN_VyQn",
        "outputId": "3c81c124-45e6-4db4-c4b1-9b93f2870e69"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(69790, 1)\n",
            "(69790, 10)\n"
          ]
        }
      ],
      "source": [
        "print(y.shape)\n",
        "print(X.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3Lupo4AcVyQo"
      },
      "source": [
        "##### KFold cross validation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Mps4gU1MVyQp"
      },
      "outputs": [],
      "source": [
        "kf = KFold(n_splits = 10, shuffle = True, random_state = 4)\n",
        "skf = StratifiedKFold(n_splits = 10, shuffle = True, random_state = 4)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "E3BNo8FLVyQp"
      },
      "outputs": [],
      "source": [
        "for train_index,test_index in kf.split(X,y):\n",
        "    X_train,X_test = X.loc[train_index],X.loc[test_index]\n",
        "    y_train,y_test = y.loc[train_index],y.loc[test_index]"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Model Training & Evaluation"
      ],
      "metadata": {
        "id": "Ye6-TuLuj9mk"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a06YxtiuVyQq"
      },
      "source": [
        "###### Logistic regression model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "QDBUd2U7VyQq"
      },
      "outputs": [],
      "source": [
        "logreg = LogisticRegression()\n",
        "grid={\"C\":np.logspace(-3,3,7), \"penalty\":[\"l1\",\"l2\"]}\n",
        "logreg_cv=GridSearchCV(logreg,grid,cv=10)\n",
        "model = logreg_cv.fit(X_train, y_train)\n",
        "y_pred = logreg_cv.predict(X_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "SvoeuL_UVyQq"
      },
      "outputs": [],
      "source": [
        "scores = logreg_cv.score(X_test,y_test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2s6HQFNrVyQr",
        "outputId": "80448f12-2dfd-49bb-cd37-62085a951f33"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of logistic regression classifier on test set: 0.85\n"
          ]
        }
      ],
      "source": [
        "print('Accuracy of logistic regression classifier on test set: {:.2f}'.format(scores))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BuGxikfWVyQr",
        "outputId": "954bd403-5058-4233-c20c-6593f840eb6b"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[2876,  566],\n",
              "       [ 469, 3068]])"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ],
      "source": [
        "confusion_matrix = confusion_matrix(y_test, y_pred)\n",
        "confusion_matrix"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IYNJYQMXVyQr",
        "outputId": "f6a9e872-a9e5-40eb-ddf2-1b34250b705f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.86      0.84      0.85      3442\n",
            "           1       0.84      0.87      0.86      3537\n",
            "\n",
            "    accuracy                           0.85      6979\n",
            "   macro avg       0.85      0.85      0.85      6979\n",
            "weighted avg       0.85      0.85      0.85      6979\n",
            "\n"
          ]
        }
      ],
      "source": [
        "print(classification_report(y_test, y_pred))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "glvGgYxYVyQr",
        "outputId": "4698d4f7-adba-4d26-ca43-55ea50056345"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "logit_roc_auc = roc_auc_score(y_test, y_pred)\n",
        "\n",
        "fpr, tpr, thresholds = roc_curve(y_test, logreg_cv.predict_proba(X_test)[:,1])\n",
        "\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(fpr, tpr, label='Logistic Regression (area = %0.2f)' % logit_roc_auc)\n",
        "plt.plot([0, 1], [0, 1],'r--')\n",
        "plt.xlim([0.0, 1.0])\n",
        "plt.ylim([0.0, 1.05])\n",
        "plt.xlabel('False Positive Rate')\n",
        "plt.ylabel('True Positive Rate')\n",
        "plt.title('Receiver operating characteristic')\n",
        "plt.legend(loc=\"lower right\")\n",
        "plt.savefig('Log_ROC')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "afbLAD91VyQr"
      },
      "outputs": [],
      "source": [
        "def logreg(X, y, cv):\n",
        "    \"\"\"\n",
        "    Creates folds manually, perform \n",
        "    Returns an array of validation (recall) scores\n",
        "    \"\"\"\n",
        "    if cv == 'kf':\n",
        "        cv = KFold(n_splits = 10, shuffle = True, random_state = 4)\n",
        "    elif cv == 'skf':\n",
        "        cv = StratifiedKFold(n_splits = 10, shuffle = True, random_state = 4)\n",
        "    else:\n",
        "        cv == None\n",
        "    \n",
        "    scores = []\n",
        "    logreg = LogisticRegression()\n",
        "    grid={\"C\":np.logspace(-3,3,7), \"penalty\":[\"l1\",\"l2\"]}\n",
        "    logreg_cv=GridSearchCV(logreg,grid,cv=10)\n",
        "    for train_index,test_index in cv.split(X,y):\n",
        "        X_train,X_test = X.loc[train_index],X.loc[test_index]\n",
        "        y_train,y_test = y.loc[train_index],y.loc[test_index]\n",
        "\n",
        "        # Fit the model on the training data\n",
        "        model_obj = logreg_cv.fit(X_train, y_train)\n",
        "        # Score the model on the validation data\n",
        "        score = recall_score(y_test, model_obj.predict(X_test))\n",
        "        scores.append(score)\n",
        "    return np.array(scores).mean()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-LMeCxGVVyQs",
        "outputId": "798411df-3eff-49e5-ce85-746282de9803"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8569456268661846"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ],
      "source": [
        "logreg(X, y, kf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u4x6sWVGVyQs",
        "outputId": "58bdfc28-e3c4-49bb-ce16-90734742f689"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8572003702179835"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ],
      "source": [
        "logreg(X, y, skf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "bfJH9QOXVyQs"
      },
      "outputs": [],
      "source": [
        "from sklearn.metrics import confusion_matrix\n",
        "def model_classifier(model, X, y, cv):\n",
        "    \"\"\"\n",
        "    Creates folds manually, perform \n",
        "    Returns an array of validation (recall) scores\n",
        "    \"\"\"\n",
        "    if cv == 'kf':\n",
        "        cv = KFold(n_splits = 10, shuffle = True, random_state = 4)\n",
        "    elif cv == 'skf':\n",
        "        cv = StratifiedKFold(n_splits = 10, shuffle = True, random_state = 4)\n",
        "    else:\n",
        "        cv == None\n",
        "    \n",
        "    scores = []\n",
        "    \n",
        "    \n",
        "    for train_index,test_index in cv.split(X,y):\n",
        "        X_train,X_test = X.loc[train_index],X.loc[test_index]\n",
        "        y_train,y_test = y.loc[train_index],y.loc[test_index]\n",
        "\n",
        "        # Fit the model on the training data\n",
        "        model_obj = model.fit(X_train, y_train)\n",
        "        y_pred = model_obj.predict(X_test)\n",
        "        # Score the model on the validation data\n",
        "        score = accuracy_score(y_test, y_pred)\n",
        "        report = classification_report(y_test, y_pred)\n",
        "        conf_matrix = confusion_matrix(y_test, y_pred) \n",
        "        \n",
        "        scores.append(score)\n",
        "        mean_score = np.array(scores).mean()\n",
        "    import seaborn as sns\n",
        "    array = conf_matrix\n",
        "    #df_cm = pd.DataFrame(array, index = ['No','Yes'], columns = ['No','Yes'])\n",
        "    plt.figure(figsize = (6,4))\n",
        "    #sns.heatmap(df_cm, annot=True,fmt='d')\n",
        "    group_names = ['True Neg','False Pos','False Neg','True Pos']\n",
        "    group_counts = ['{0:0.0f}'.format(value) for value in array.flatten()]\n",
        "    group_percentages = ['{0:.2%}'.format(value) for value in array.flatten()/np.sum(array)]\n",
        "    labels = [f\"{v1}\\n{v2}\\n{v3}\" for v1, v2, v3 in zip(group_names,group_counts,group_percentages)]\n",
        "    labels = np.asarray(labels).reshape(2,2)\n",
        "    sns.heatmap(array, annot=labels, fmt='', cmap='Blues')\n",
        "    print('\\n Accuracy scores of the model: {:.2f}'.format(mean_score))\n",
        "    print('\\n Classification report of the model')\n",
        "    print('--------------------------------------')\n",
        "    print(report)\n",
        "    \n",
        "    print('\\n Confusion Matrix of the model')\n",
        "    print('--------------------------------------')\n",
        "    #print(conf_matrix)\n",
        "    \n",
        "    \n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Bo15L8mGVyQs"
      },
      "outputs": [],
      "source": [
        "logreg = LogisticRegression()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 604
        },
        "id": "QxX7MUOlVyQt",
        "outputId": "e0767738-bf3f-4d43-be5d-e05320b800cb"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            " Accuracy scores of the model: 0.84\n",
            "\n",
            " Classification report of the model\n",
            "--------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.86      0.84      0.85      3442\n",
            "           1       0.84      0.87      0.86      3537\n",
            "\n",
            "    accuracy                           0.85      6979\n",
            "   macro avg       0.85      0.85      0.85      6979\n",
            "weighted avg       0.85      0.85      0.85      6979\n",
            "\n",
            "\n",
            " Confusion Matrix of the model\n",
            "--------------------------------------\n",
            "CPU times: user 2.34 s, sys: 1.52 s, total: 3.86 s\n",
            "Wall time: 3.03 s\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "%%time\n",
        "model_classifier(logreg, X, y, kf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 572
        },
        "id": "Lf8NrW56VyQt",
        "outputId": "e43a4b15-39d7-4b04-9fae-59b3fb2a48c6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            " Accuracy scores of the model: 0.84\n",
            "\n",
            " Classification report of the model\n",
            "--------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.86      0.84      0.85      3489\n",
            "           1       0.84      0.87      0.85      3490\n",
            "\n",
            "    accuracy                           0.85      6979\n",
            "   macro avg       0.85      0.85      0.85      6979\n",
            "weighted avg       0.85      0.85      0.85      6979\n",
            "\n",
            "\n",
            " Confusion Matrix of the model\n",
            "--------------------------------------\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "model_classifier(logreg, X, y, skf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "AxWQay-EVyQt"
      },
      "outputs": [],
      "source": [
        "xgb = XGBClassifier()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 604
        },
        "id": "wIg9sO8iVyQt",
        "outputId": "7338835a-e85d-4d91-f55c-898f89810289"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            " Accuracy scores of the model: 0.87\n",
            "\n",
            " Classification report of the model\n",
            "--------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.91      0.84      0.87      3442\n",
            "           1       0.86      0.92      0.89      3537\n",
            "\n",
            "    accuracy                           0.88      6979\n",
            "   macro avg       0.88      0.88      0.88      6979\n",
            "weighted avg       0.88      0.88      0.88      6979\n",
            "\n",
            "\n",
            " Confusion Matrix of the model\n",
            "--------------------------------------\n",
            "CPU times: user 46.6 s, sys: 132 ms, total: 46.7 s\n",
            "Wall time: 46.6 s\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "%%time\n",
        "model_classifier(xgb, X, y, kf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 569
        },
        "id": "jHUxkjlCVyQt",
        "outputId": "dbed55f7-c5b1-4091-a8a9-78ec0e861917"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            " Accuracy scores of the model: 0.87\n",
            "\n",
            " Classification report of the model\n",
            "--------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.91      0.83      0.87      3489\n",
            "           1       0.85      0.91      0.88      3490\n",
            "\n",
            "    accuracy                           0.87      6979\n",
            "   macro avg       0.88      0.87      0.87      6979\n",
            "weighted avg       0.88      0.87      0.87      6979\n",
            "\n",
            "\n",
            " Confusion Matrix of the model\n",
            "--------------------------------------\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "model_classifier(xgb, X, y, skf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gBhWnJuzVyQt"
      },
      "outputs": [],
      "source": [
        "mlp = MLPClassifier()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 604
        },
        "id": "3LXDx13PVyQt",
        "outputId": "bf63db42-4cc7-46bb-c15f-a29c85423bd2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            " Accuracy scores of the model: 0.90\n",
            "\n",
            " Classification report of the model\n",
            "--------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.95      0.86      0.90      3442\n",
            "           1       0.87      0.95      0.91      3537\n",
            "\n",
            "    accuracy                           0.91      6979\n",
            "   macro avg       0.91      0.91      0.91      6979\n",
            "weighted avg       0.91      0.91      0.91      6979\n",
            "\n",
            "\n",
            " Confusion Matrix of the model\n",
            "--------------------------------------\n",
            "CPU times: user 14min 13s, sys: 11min 23s, total: 25min 37s\n",
            "Wall time: 13min 12s\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "%%time\n",
        "model_classifier(mlp, X, y, kf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 569
        },
        "id": "Lhe69_SdVyQt",
        "outputId": "2a343791-b8a4-427b-cde6-191865673bc8"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            " Accuracy scores of the model: 0.89\n",
            "\n",
            " Classification report of the model\n",
            "--------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.94      0.86      0.90      3489\n",
            "           1       0.87      0.95      0.91      3490\n",
            "\n",
            "    accuracy                           0.91      6979\n",
            "   macro avg       0.91      0.91      0.91      6979\n",
            "weighted avg       0.91      0.91      0.91      6979\n",
            "\n",
            "\n",
            " Confusion Matrix of the model\n",
            "--------------------------------------\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "model_classifier(mlp, X, y, skf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "CqC2zGmwVyQt"
      },
      "outputs": [],
      "source": [
        "clf = svm.SVC(kernel='linear')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 604
        },
        "id": "Bddc2gJEVyQu",
        "outputId": "0fa32245-8017-4599-d39e-6fec1dd9da3c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            " Accuracy scores of the model: 0.85\n",
            "\n",
            " Classification report of the model\n",
            "--------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.88      0.82      0.85      3442\n",
            "           1       0.84      0.89      0.86      3537\n",
            "\n",
            "    accuracy                           0.86      6979\n",
            "   macro avg       0.86      0.86      0.86      6979\n",
            "weighted avg       0.86      0.86      0.86      6979\n",
            "\n",
            "\n",
            " Confusion Matrix of the model\n",
            "--------------------------------------\n",
            "CPU times: user 18min, sys: 2.66 s, total: 18min 3s\n",
            "Wall time: 18min\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "%%time\n",
        "model_classifier(clf, X, y, kf)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QOFiCTy7VyQu"
      },
      "source": [
        "###### ROC plots "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Lpxgw08CVyQu"
      },
      "outputs": [],
      "source": [
        "def roc_plot(model, X, y, cv):\n",
        "    for train_index,test_index in cv.split(X,y):\n",
        "        X_train,X_test = X.loc[train_index],X.loc[test_index]\n",
        "        y_train,y_test = y.loc[train_index],y.loc[test_index]\n",
        "        model_obj = model.fit(X_train, y_train)\n",
        "        y_pred = model_obj.predict(X_test)\n",
        "        y_pred_prob = model_obj.predict_proba(X_test)[:,1]\n",
        "        \n",
        "        logit_roc_auc = roc_auc_score(y_test, y_pred)\n",
        "        fpr, tpr, thresholds = roc_curve(y_test, y_pred_prob)\n",
        "    plt.figure()\n",
        "    val_model = input(\"Enter your model name: \")\n",
        "    plt.plot(fpr, tpr, label= val_model + ' (area = %0.2f)' % logit_roc_auc)\n",
        "    plt.plot([0, 1], [0, 1],'r--')\n",
        "    plt.xlim([0.0, 1.0])\n",
        "    plt.ylim([0.0, 1.05])\n",
        "    plt.xlabel('False Positive Rate')\n",
        "    plt.ylabel('True Positive Rate')\n",
        "    plt.title('Receiver operating characteristic')\n",
        "    plt.legend(loc=\"lower right\")\n",
        "    my_fig = val_model + '.png'\n",
        "    #plt.savefig(os.path.join(my_path, my_fig))\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "PG_evzCQVyQu",
        "outputId": "b2925a48-712e-4cdc-98d4-ed34d0867780"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Enter your model name: Logistic Regression \n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "roc_plot(logreg, X, y, kf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "tUB9YXe8VyQu",
        "outputId": "d2460657-1f5a-4c42-bd5e-56708e4a84c0"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Enter your model name: Logistic Regression skf\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "roc_plot(logreg, X, y, skf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "6RW8N6DJVyQu",
        "outputId": "340d70b2-ff47-48c7-83e1-f11170afc1be"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Enter your model name: XGBoost\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "roc_plot(xgb, X, y, kf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "iBzabbz7VyQu",
        "outputId": "f8b509e2-75a7-48fa-c9da-9c3cac3d4f1a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Enter your model name: XGBoost skf\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "roc_plot(xgb, X, y, skf)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "v-32WMskVyQu",
        "outputId": "8d64fc86-1714-4efe-f04f-d09d1bdbed99"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Enter your model name: MLP\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "roc_plot(mlp, X, y, kf)"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "0n3jEfmRarvr"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.6"
    },
    "toc": {
      "base_numbering": 1,
      "nav_menu": {},
      "number_sections": true,
      "sideBar": true,
      "skip_h1_title": false,
      "title_cell": "Table of Contents",
      "title_sidebar": "Contents",
      "toc_cell": false,
      "toc_position": {},
      "toc_section_display": true,
      "toc_window_display": false
    },
    "colab": {
      "provenance": [],
      "toc_visible": true,
      "include_colab_link": true
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}